植物园室外场景定位与建图标准数据集

主要面向自主机器人与人工智能研究、高质量多传感器SLAM数据集需求建设，基于自主研发的一体式多传感器平台采集产生，基于双目彩色相机、双目灰度相机、激光雷达、轮速计、IMU等传感器进行高时空精度的同步采集，采集场景为上海交通大学植物园，场景特点为室外环境不稳定，光照、纹理复杂。存在重复纹理、复杂结构、遮挡、视角变化大等。

This dataset is constructed to address the demand for high-quality multi-sensor SLAM datasets for autonomous robotics and artificial intelligence research. It is collected using an independently developed integrated multi-sensor platform, which synchronously acquires data with high spatiotemporal accuracy via multiple sensors including binocular color cameras, binocular grayscale cameras, LiDAR, wheel odometers, and IMUs. The data collection was conducted at the Shanghai Jiao Tong University Botanical Garden, where the outdoor environment is unstable with complex lighting and textures, and contains challenges such as repetitive textures, complex structures, occlusions, and significant viewpoint variations.

搜集汇总

背景与挑战

背景概述

该数据集是一个面向自主机器人与人工智能研究的多传感器SLAM标准数据集，主要用于室外场景的定位与建图。它基于自主研发的一体式多传感器平台采集，包含双目彩色相机、双目灰度相机、激光雷达、轮速计和IMU等传感器的高时空精度同步数据，数据量达91.09GB，格式包括.bag、.zip和.rar。采集场景为上海交通大学植物园，具有室外环境不稳定、光照纹理复杂、重复纹理、复杂结构、遮挡和视角变化大等特点，适用于处理复杂视觉挑战的算法开发和评估。